Skip to main content

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: Listening tests with Apple Sound Check (Read 4962 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Listening tests with Apple Sound Check

I'm doing some very informal AB (not ABX) testing of albums to compare the effects of the "loudness wars" on perceived quality of music and dynamic range.

So, here is what I did.  I grabbed 2  copies of the Electric Light Orchestra album "A New World Record," one CD released in 1990 and one released in 2006 with bonus tracks.  I ripped them both to Apple Lossless and dropped them into iTunes and let iTunes Calculate Sound Check data for all the songs.

I am playing all the songs in iTunes over an AirPort Express using AirPlay to my stereo.  None of this should matter, but I thought i should point this out just for full disclosure.

I have no way to scientifically measure the loudness of what's coming out of my speaker, so all observations are purely what I am hearing.

Here's is what I got.

Looking at the first track on the CD, TightRope, Sound Check says the following:

2006 release:  -7.02 dB
1990 release:  +0.5 dB

So, it's clear that the 2006 remaster has been compressed, as is the norm these days with remasters, sionce louder sounds better.

So, I go to play back the 2006 tightrope and let it go 15 seconds and then hop onto the 1990 tightrope.

I am pre-disposed to assuming that the 2006 remaster will sound worse, and to my skewed brain that already didn't like it before I even listened to it, it was worse.  But it was worse for an odd reason.  With Sound Check enabled, the 1987 release sounds significantly louder than the 2006 release.  I can confirm that when I turn off Sound Check in the iTunes preferences, the 2006 release is definitely louder that the 1987 release, so I know Sound Check is being applied.

This has me asking a couple of questions.

1. Is there something in the way music is dynamically compressed that makes things like Sound Check get skewed.  Does dynamic ranger effect Sound Check accuracy?

2. Is there any way to stream via AirPlay with an app that support ReplayGain?  I'd love to compare Sound Check to ReplayGain and see if it has the same supposed issue with the older remaster sounding louder.

I repeated all this with the ELO album Balance of Power, and compared 1986 to 2007 and had similar observations, with the Sound Check enabled in the preferences causing the older release to appear louder.

Then my lunch hour was over and I had to stop for now.

Re: Listening tests with Apple Sound Check

Reply #1
2. Check out these options:
https://rogueamoeba.com/airfoil/
http://emilles.dyndns.org/software/out_apx.html

1. Consider that applying dynamic range compression amounts to tweaking the volume knob as the track plays (possibly multiple volume knobs for different EQ bands).. it makes sense that you're not going to be able to restore the original volume levels by applying one gain adjustment to the whole track. So, I think it's expected that SoundCheck/replaygain are unable to give you the same perceived volume level for the uncompressed and compressed versions of the track.

Re: Listening tests with Apple Sound Check

Reply #2
2. Check out these options:
https://rogueamoeba.com/airfoil/
http://emilles.dyndns.org/software/out_apx.html

1. Consider that applying dynamic range compression amounts to tweaking the volume knob as the track plays (possibly multiple volume knobs for different EQ bands).. it makes sense that you're not going to be able to restore the original volume levels by applying one gain adjustment to the whole track. So, I think it's expected that SoundCheck/replaygain are unable to give you the same perceived volume level for the uncompressed and compressed versions of the track.

Would the less compressed track sound louder when SC/RG is applied since it has a greater difference between loud and quiet parts of the song?

Re: Listening tests with Apple Sound Check

Reply #3
My first guess would be "loudness is complicated"...   If you have a compressed recording and a dynamic recording and you asked two people to adjust the volume to make them equally loud, you might get two different results.

When I read the description of Sound Check it seems to imply some limiting, and if so that's a different approach than ReplayGain.

Quote
2. Is there any way to stream via AirPlay with an app that support ReplayGain? 
You can use WaveGain (or MP3Gain) to "permanently" change the volume of the files themselves.   It would be intersting to know if the ReplayGain algorithm is better than Sound Check or vice-versa, although one or two samples doesn't "prove" than one or the other is better overall.

Quote
I am pre-disposed to assuming that the 2006 remaster will sound worse, and to my skewed brain that already didn't like it before I even listened to it, it was worse.  But it was worse for an odd reason. 
You might try volume matching yourself, by ear, and then compare the two versions.    (I assume you have an audio editor so you can make volume-adjusted versions of the songs?)

And, there may be some EQ or noise reduction or other effects applied during re-mastering, so compression/limiting probably aren't the only differences.